Representativeness of geometrical details during wind tunnel tests. Application to train aerodynamics in crosswind conditions

This work aims to highlight the influence of the scale of reproduction of geometric details on models during wind tunnel tests. An application is presented concerning the reproduction of high voltage lines present on the roof of trains. It is shown that the presence of these geometric artefacts deeply modifies the forces and moments measured under cross-wind. It is also shown that the reproduction of geometric details with the scale of the model is not a sufficient condition to ensure a good representativeness of the tests. It is necessary to ensure that the changes in the flow dynamics caused by the presence of the details are correctly reproduced (separation-reattachment process on the roof of the train in this study). The influence of the scale of reproduction of the geometrical details is also studied and a method is proposed to extrapolate the results obtained without roof-lines in order to predict the rolling moment in the presence of roof-lines.